Fig 2 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Collaborative filtering vs. content-based approach
Source publication
Recommender systems are widely used in various applications, such as online shopping, social media, and news personalization. They can help systems by delivering only the most relevant and promising information to their users and help people by mitigating information overload. At the same time, algorithmic recommender systems are a new form of gate...
Context in source publication
Context 1
... gives rise to significant concerns regarding privacy and misuse. RS can face a challenge when they encounter new users since there may not be enough data to build a user model. This situation is called "cold-start problem," and the best way to address it depends on the specific use case. The same is true for items that are new to the system. In Fig. 2, the two most common RS approaches, i.e., collaborative filtering and the content-based approach, are conceptually shown. Collaborative filtering encounters the cold-start problem when there is a lack of user-item interaction data, making it challenging to identify similar users. For new users, the recommender system may suggest ...
Similar publications
Citations
... The idea that personal values are well-rooted, stable and defined concepts, and take into account the neuroplasticity of the brain seems to lose its validity with users immersed within social media algorithms, which may impact it at both the individual and group levels. This context raises complex ethical questions and debates about the long-term societal effects but also individual effects on privacy, consent, personal identity and fairness, which are at the advent of their exploration among social sciences and mental health professionals with a paucity of studies examining how these algorithms interact with and influence personal values [11,12,[35][36][37]. ...
The current study contributes to the literature by assessing the associations between personal values, explored with Schwartz`s Portrait Values Questionnaire, social media behaviors, assessed with Bergen Scale of Social Media Addiction and Social Media Motivations to Use Scale and psychological well-being assessed with Patient Health Questionnaire-4, in a sample of first-year medical students. It was examined medical students’ personal values profiles and the perceived influence of social media on self-aspects. All participants (N = 151) were Romanian and English module students, young (average age = 19.2, SD = 1.5), 68,9% females and 31,1% males. Pearson coefficient correlation analyses were performed to verify the associations between the main four clusters of personal values (Self Transcendence, Self Enhancement, Openness to change and Conservationism) with social media behaviors and psychological wellbeing. The most frequent cluster of values was Self-Transcendence (M = 5.21) while the least was Self-Enhancement (M = 4.05). There was no significant correlation between social media addiction, psychological wellbeing and a specific cluster of values while the perception of self-aspects influenced by social media included involvement in community problems, creativity for Openness to change group (R = .24;.22, p < .05), tolerance towards sexual minorities and self-evaluation in Self Transcendence group (R = .24;.21;.42, p < .05) while Conservationism and Self Enhancement groups didn`t report any change. The findings highlight the need for awareness and education of medical students and general population in the field of Digital Ethics including social media complex impact on personal values as AI-algorithms may imply a potential destabilization and perpetual shaping of one`s behavior with still unpredictable individual and societal effects.
... Moreover, psychological factors such as personality traits, emotions, and social connections significantly influence decision-making outcomes. Consequently, these factors feature prominently in literature discussing the incorporation of psychological aspects into recommender system development, particularly concerning users' personality, emotions, and decision-making behavior [3,12,25]. Notably, trust has been identified as a significant factor in enhancing recommendation processes [27]. Recent research on trust in RSs shows that there are multiple ways that trust can be influenced [14]. ...
... This concept has garnered attention for its potential to mitigate popularity bias and enhance recommendation utility by promoting better discoverability. However, achieving serendipity poses challenges, as it necessitates striking a balance between surprise and relevance [5,12,22]. Recent insights highlight the close relationship between novelty and serendipity, both evaluating recommended items based on a user's historical interactions and emphasizing the discovery of content that aligns with personal preferences [6]. Prioritizing diversity and serendipity not only enriches user experiences but also fosters fairness by ensuring a more equitable distribution of recommendations across items, mitigating the tendency of RSs to favor popular items consistently [17]. ...
... Investigating how enhanced RSs ensure inclusivity, representation, and ethical alignment with societal values is essential [14]. Grounding this research in responsible AI principles ensures the development of ethically sound and inclusive systems that foster informed user experiences [1] in order to create RSs aligned with human values and well-being [12,24]. ...
